Association between parental body mass index and autism spectrum disorder: a systematic review and meta-analysis.

Studies have examined the association between parental body mass index (BMI) and autism spectrum disorder (ASD) in offspring, with inconsistent results, especially regarding maternal obesity, overweight and underweight. Cochrane Library, EMBASE, PubMed and PsycINFO databases were searched up to March 2018 for relevant observational studies with no language restriction. Our literature search identified 13 eligible studies for meta-analysis (involving 943,293 children and 30,337 cases). For maternal BMI (13 studies), both maternal obesity [OR 1.41 (95% CI 1.19-1.67)] and maternal overweight [OR 1.16 (95% CI 1.05-1.27)] were significantly associated with ASD, while maternal underweight was not associated with ASD [OR 1.08 (95% CI 0.98-1.20)]. For paternal BMI (three studies), no association was found (paternal obesity: OR 1.28, 95% CI 0.94-1.74; overweight: OR 1.07, 95% CI 0.99-1.15; underweight: OR 1.12, 95% CI 0.87-1.44). Pooled estimates were robust in sensitivity analysis and subgroup analyses. Publication bias may exist for studies assessing maternal BMI and ASD risk, but the filled estimates were not altered. Relative to normal weight, maternal obesity and overweight were significantly associated with increased ASD risk, while maternal underweight was not associated with ASD. Although no association between paternal BMI and ASD was found, current evidence is limited (three studies). Future studies are warranted to address more confounding factors and to identify potential mediators of the association, but pre-pregnancy weight control is suggested.